Estimation Apparatus, Estimation Method, and Non-Transitory Computer-Readable Storage Medium for Storing Estimation Program

1. A non-transitory computer-readable storage medium for storing an estimation program which causes a processor to perform processing for object recognition, the processing comprising: executing learning processing by training an autoencoder with a data group corresponding to a specific task; calculating a degree of compression of each part regarding data included in the data group by using the trained autoencoder; and estimating a common part with another piece of data included in the data group regarding the data corresponding to the specific task based on the calculated degree of compression of each part.

2. The non-transitory computer-readable storage medium according to claim 1 , wherein the learning learns the autoencoder in such a manner that an input and an output of the autoencoder correspond with each other by using the data group which corresponds to the specific task and is unsupervised, and the calculating generates a first intermediate feature obtained by inputting data included in the data group and causing the data to be learned by using the autoencoder, and generates a second intermediate feature obtained by inputting data resulting from processing a part of the data and causing the data to be learned, and calculates a degree of compression of each part of the data by using the first intermediate feature and the second intermediate feature.

3. The non-transitory computer-readable storage medium according to claim 1 , wherein the estimating generates an auxiliary image in which a part with a higher compression rate is highlighted to a larger extent regarding the data group by using the degree of compression of each part and estimates the common part based on the auxiliary image generated.

4. The non-transitory computer-readable storage medium according to claim 3 , wherein the estimating further divides target data into regions by a given method and estimates a region about which a compression rate is higher than a given value as the common part from the auxiliary image by using the divided regions.

5. The non-transitory computer-readable storage medium according to claim 3 , wherein the estimating further divides target data into regions by a given method, and calculates an average of the degree of compression obtained from the auxiliary image regarding each divided region, and couples divided regions based on the average of each divided region, and estimates the coupled regions as the common part.

6. The non-transitory computer-readable storage medium according to claim 1 , wherein the data group is a data group of image data, a data group of audio data, or a data group of text data.

7. An estimation apparatus comprising: a memory; and a processor coupled to the memory, the processor being configured to execute a learning processing that includes training an autoencoder with a data group corresponding to a specific task; execute a calculating processing that includes calculating a degree of compression of each part regarding data included in the data group by using the autoencoder learned by the learning processing; and execute an estimating processing that includes estimating a common part with another piece of data included in the data group regarding the data corresponding to the specific task based on the degree of compression of each part calculated by the calculating processing.

8. An estimation method implemented by a computer, the estimation method comprising: executing learning processing by training an autoencoder with a data group corresponding to a specific task; calculating a degree of compression of each part regarding data included in the data group by using the trained autoencoder; and estimating a common part with another piece of data included in the data group regarding the data corresponding to the specific task based on the calculated degree of compression of each part.